The invention is in the field of construction, and more particularly relates to securely fastening together wide expanses of wood cross-grain, in which configuration shrinkage of the wood over time creates slack along the bolt which spans the shrinking members.
The invention particularly relates to protection of structures in high winds or an earthquake. Over the last few decades, considerable professional attention has been devoted studying various aspects of earthquakes such as how quakes can be predicted, how to design high-rise office buildings on float plates so that they will not feel the full force of the horizontal jolts that earthquakes produce, and how to modify buildings and other structures to minimize the danger to those inside and nearby. Much effort has gone into the development of construction techniques which minimize injuries caused by falling glass and collapsing building structures.
This effort has been devoted largely to mitigating the human cost, in terms of lives and serious injury, caused by a Big One. Although these considerations are of primary concern, for every major earthquake which seriously threatens life and property there are many quakes which shake buildings and cause considerable cosmetic damage even though none of the buildings are significantly impaired from a structural standpoint.
The instant invention addresses this type of damage which is largely cosmetic, and focuses on a particular detail of current construction which results in instability in the framing of wallboard, plaster, brick veneer or other surface panels, which in turn causes the overstressing of these panels at frame attachment points in extreme conditions of wind or earthquake. These panels are strong and can withstand considerable shock without damage if on secure framing.
However, when framed above an underlying horizontal wood support member such as a joist which has a thickness exceeding several inches, trauma damage is particularly likely. This is due to the fact that hold-down bolts which pass vertically through the structural members beneath the panels may span a wood "sandwich" of 10 to 15 inches normal to the grain, and considerable axial play generally develops between the bolt head and the retaining nut over time, as the wood dries out and shrinks. These bolts, commonly used for both second stories and for the first story if the house is built on a foundation wall, hold down the bottom of the frame, and it is the frame which prevents the laterally-bracing shear walls from overturning.
These long bolt spans may create slack up to 3/8ths inch or even 1/2 an inch in an extreme case. In a quake, panels overlying these areas are readily damaged as the frame distorts in shape as it raises off of the sill plate at one or both ends. The framing "rattles" around, causing stress at the drywall joining points, resulting in fracture lines and segments torn loose from the wall.
Beyond cosmetic damage, earthquake motion causes a dynamic "hammering" effect on the building, which multiplies the forces on the building structure to levels not considered in structural engineering calculations when the structure was designed. Premature structural failure may occur. If a three-foot by eight-foot shear wall is allowed to have one corner uplift by 3/8 inch due to slack in the hold-down member, the lateral displacement (or "drift" as it is called in the building code) of the top will be about one inch, which is double the allowable drift set forth in the Code. Current construction practices do not address this wood shrinkage problem because there is no readily available solution.
In an instance in which long vertical bolts are used, if the connection of the bottom of the wallboard framing to the underlying structure were modified to eliminate the possibility of subsequently developed slack along the bolt, much of this damage could be avoided. The problem caused by wood shrinkage as it relates to maintaining a tight connection is addressed in U.S. Pat. No. 4,812,096, disclosing a SELF-TIGHTENING NUT. This invention involves log cabin wall logs wherein each log is fastened to the log underneath with long nails which pass entirely through the upper log and sink into the lower one. The invention is a long nail with a coil spring under the head so the nail is continuously tensioned and compensates for shrinkage.
Different considerations than those in the log cabin log fastener present themselves regarding frame bolt adjustment for quake resistance, as set forth in U.S. Pat. No. 4,812,096 issued Mar. 14, 1989 on a SELF-TIGHTENING NUT. The device of this invention addresses the problem head-on. It incorporates a long coiled spring capable of considerable rotational travelling while exerting considerable torque mounted in a housing over the nut. The spring is wound up much like a window shade coil spring, and as slack occurs along the bolt, the nut is automatically tightened by the coil spring.
This approach is very direct. The coil spring mechanism performs the function of a workman on-site, who would tighten the bolt down periodically as it becomes loose. However, even though it is a direct assault on the problem and stands as testimony to the existence of the problem, it is heavy-handed in its approach, using a long, bulky housing cylinder which is longer than the length of the bolt itself due to the number of winds in the coil spring that are necessary. Even more serious is a possibility of failure. The unit would be out of sight, ordinarily inside a wall, and even if it were in the open, it is questionable whether its non-operative condition would be noticed. A nut that has been on a bolt for ten, twenty, or thirty years is not going to be easy to turn. It is very doubtful that such a nut could be rotated by a coil spring.
There is a need for a self-tightening bolt construction that is simple, compact and substantially fool proof, so that it can be out of sight, hidden within walls for years, and still function properly substantially without fail. Because of the nature of the problem addressed, for every 1000 that are installed only a very few would every be tested by an earthquake, and that may not happen for years after installation, making marketplace feedback a poor tool for improving reliability. Instead, the design must be inherently foolproof from the outset.